© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 0-GEMMASCOPE-TRANSCODER-16K
    7. 11623
    Prev
    Next
    INDEX
    Explanations

    single letters, often capitalized, and often followed by numbers or other symbols

    oai_token-act-pair · gemini-2.0-flash

    code snippets/formulas

    np_max-act · gemini-2.0-flash

    uppercase letters Method used: 1 — reason: all max tokens are single capital letters (G, F, E, H, B)

    np_max-act-logits · gpt-5-miniTriggered by @chenshw0109
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_0/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.0.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Anſ
    -1.70
     Houſe
    -1.65
     myſelf
    -1.62
     Reſ
    -1.56
     greateſt
    -1.55
     ſever
    -1.55
     purpoſe
    -1.55
     Diſ
    -1.55
     houſe
    -1.54
     Perſ
    -1.46
    POSITIVE LOGITS
     B
    1.42
     D
    1.41
     S
    1.41
     P
    1.40
     F
    1.39
     C
    1.37
     H
    1.32
     M
    1.30
     R
    1.30
     T
    1.30
    Activations Density 3.236%

    No Known Activations